Most work in the area of brain disease research focuses on neuronal mechanisms of toxicity. However, glia comprise the most abundant class of cells in the nervous system and subserve important roles in neurotransmitter uptake, ammonia detoxification and other critical processes. Although astrocytes appear to be in a primary position to affect brain function in pathological conditions, little is known about the functional role of these cells in disease states. Mutations in the gene encoding the astrocyte-specific intermediate filament, glial fibrillary acidic protein (GFAP), cause Alexander disease, a typically childhood disorder that manifests with seizures and severe white matter pathology. Dysmyelination is accompanied by the formation of GFAP-rich inclusions in astrocytes termed Rosenthal fibers. The presence of GFAP in these aggregates and the observation that overexpression of GFAP in mouse astrocytes produces a severe neurological syndrome and Rosenthal fiber formation has led to the hypothesis that Alexander disease is produced by a dominant gain of function mechanism, perhaps related to abnormal aggregation of GFAP. To test this hypothesis and create a1 model of Alexander disease, we have expressed normal and mutant versions of GFAP in Drosophila. We find that overexpression of GFAP in the Drosophila retina and glia leads to formation of numerous GFAP-containing, Rosenthal-fiber like inclusion bodies. Inclusion body formation is accompanied by degeneration in the retina. In this pilot proposal we will optimize the retinal degeneration phenotype for use in genetic screening experiments. We will then perform a limited preliminary screen to evaluate the utility of the retinal degeneration phenotype and to identify genetic modifiers of GFAP toxicity. Relevance: Although nonneuronal (or glial) cells are important in normal brain function, their role in disease states is largely unknown. We have created a model a rare, devastating disorder called Alexander disease, to explore the role of glia in neurological diseases. A prominent feature of Alexander disease, and of our fruit fly model of the disoder, is the formation of insoluble protein aggregates. More common disorders like Alzheimer's disease and Parkinson's disease also have abnormal protein aggregates. Thus, our work may have important implications for understanding and therapy of these diseases as well.